MIT researchers say GPT-6 Astra edited 0.016% of Qwen3.5-0.8B weights without training, fixing reasoning loops and lifting MATH-500 from 7.9% to 48.1%.

MIT researchers just gave Alibaba's Qwen AI model a serious math upgrade using GPT-6 Astra. By tweaking only 0.016% of its neural network, Astra helped Qwen ditch repetitive thinking and solve ...

In the strongest example of "weight space programming", as the technique is called, Astra’s intervention raised Qwen3.5-0.8B’s MATH-500 score from 7.9% to 48.1% while touching approximately 0.016% of ...

MIT CSAIL researchers say GPT-6 Astra edited 0.016% of Qwen3.5-0.8B weights to fix reasoning loops, lifting MATH-500 from 7.9% to 48.1% without training.

A Python script to audit and patch high-risk unary negation precedence bugs in Excel formulas. - dfrykberg/excel-unary-negation-fix ...

Article by Shenchao TechFlowNo training required, no data needed, no gradient descent.Researchers at MIT fed the weights of ...

MIT has unveiled an innovative, groundbreaking paradigm dubbed "Weight Space Programming", representing a transformative new ...

Machine learning is often used to predict future values from data. But what if, instead of just predictions, you could ...

DEMATELab is a new open-source Python application that unites classical, fuzzy, grey and Z-number DEMATEL analysis with ...

Timbral Dynamics at the Instability Boundary.” It looked at the exact physical factors that let string players make a warm, ...

That is the basic idea behind bond pricing. A bond is a promise of future payments, and to know what that promise is worth ...

OpenAI said in an announcement on October 26, 2026, that its internal models produced three research results in mathematics, ...